Development of an Emission Development of an Emission Reduction Term for NearReduction Term for Near--Source Source

DepletionDepletionChattenChatten Cowherd Cowherd

Mary Ann GrelingerMary Ann GrelingerMidwest Research InstituteMidwest Research Institute

Dick L. GebhartDick L. GebhartERDCERDC--CERLCERL

May 17 2006May 17 2006

Background

Air Quality Models for Fugitive Dust Consistently Overpredict Ambient PM-10 Concentrations by up to Factor of 4

Overprediction Places Unnecessary Restrictions on Military Training Activities

Military Training Activities

Unpaved RoadsTank TrailsVehicle ManeuveringConstruction ActivitiesSmoke Releases

Special Characteristics of Fugitive Dust

Ground Level Release with Electrostatic TribochargingWide Variety of Particle Sizes, from Sub-Micron to > 100 Micron DiametersAt Point of Release, Most Dust Mass Confined to Height of Vehicle or Other Source of Energy Input

Elevated Particle Capture Surfaces

TreesOther VegetationBuilding StructuresRough Terrain

Fugitive PM Emissions

Total PMMeasured at 5 m from source

Depleted PMCaptured within 100 m from source through variety of depletion mechanisms

Transportable PMLofted for transport beyond 100 mUnaffected by non-gravitational forces

Roadway Plume Profiling

Concentration Wind speed Mass flux Deposition Chemical constituents

Sampling SchemesFixed towers—vertical arraysMobile monitors (continuous/integrating) Remote sensing (e.g., FTIR)Sampler Deployment

Edge of source 20 to 50 m downwind

Traffic (Roadways)Freely flowingCongested

Open Source Characterization

Measurement of plume concentration, wind speed and mass flux profiles at the edge of the source and at points downwindMeasurement of particle deposition (plume depletion) vs. distance from sourceDetermination of plume losses on vegetation and other types of groundcover

Collocated Plume Profilers

On-board Mobile Monitor

MRI Recirculating Wind Tunnel

Uniform Wind Velocity (< 10% variation over working cross section )Uniform Dust Concentrations (< 10% variation)Steady State Conditions (~ 2 hr) PM-10 Concentrations (up to 10 mg/m3

using 20 l/min injection air flow)

Exposure Chamber

withSampler

Inlets

MRI Laboratory Wind Tunnel

Viewing Window with

Continuous PM Mass Monitor

Ref. Method Sampling Equipment

Vegetative Porosity: Wind Speed Reduction in Forests

Vegetative Capture of ParticlesTwo Series of Tests at Ft. Leonard Wood, MO, 2003-2004Two Series of Tests at Ft. Riley, KS, 2005

MRIStreamlined Profiling Tower in Trees

Collocated HVS-3 Cyclones

Deposition on Tall Grass at Ft. Riley KS

Example Plume Profiles

Type of Vegetation PM-10 Plume Loss*

Short Grass < 10%

Tall Grass 35-45%

Tall Cedar Trees 45-67%

Short Cedar Trees 29%

Tall Oak Trees (light winds)

41-50%

*20-30 m travel distance

PM-10 Deposition

Capture Efficency vs Wind Speed (h=3m)y = 5.8604x + 17.042

R2 = 0.8914

0

10

20

30

40

50

60

70

0 2 4 6 8 10

Wind Speed (mph)

Cap

ture

Effi

cenc

y

DFS Short GrassDFS Cedar TreesFLW Oak TreesFLW Cedar Trees

PM-10 Capture Efficiency on Oak and Cedar Trees

PM-10 Capture Efficiency on Tall Grass

Capture Efficency vs Travel Time

0

5

10

15

20

25

30

35

40

45

50

0 2 4 6 8 10 12

Trave Time to Tower B (s)

Cap

ture

Effi

cenc

y (%

)

Site 1 - 0.3 m Prarie GrassSite 2 - 1.2 m Prarie Grass

PM2.5/PM10 Ratios as Function of Downwind Distance (Oak Trees)

PM-2.5 / PM-10 Ratio vs Travel Distance

0.00

0.02

0.04

0.06

0.08

0.10

0.12

0.14

0.16

0.18

0.20

0 5 10 15 20 25 30

Distance (m)

PM

-2.5

/ PM

-10

Ratio

Ft. Leonard Wood (Trees)

Options for Modeling Representation of Near-Source DepletionMitigation Term on Emission Factor

Similar to any other natural mitigation effectUncontrolled emission factor remains in tact

Improved Deposition Algorithm within Dispersion Model

Builds on past approach to treatment of deposition

Mitigation Term--Advantages

Does not require changes to EPA modelsNear-source deposition zone (< 50-100 m) not normally associated with reliable predictionsVegetation (groundcover) is placed in same category as other dust control measures

Mitigation Term--Disadvantages

Near-source concentrations will be grossly underpredictedFor local effects, term must be direction specific, unless groundcover is uniformMay cause confusion over permitted emissions

Improved Algorithm--Advantages

Correctly represents actual emissions at the sourceHas the potential to correctly represent near-source plume impactsIs consistent with current algorithms that represent plume loss phenomena

Improved Algorithm--Disadvantages

Near source depletion phenomena will require complex mathematical representationWill also require detailed near-source groundcover inputsEPA model changes will require lengthy review period

Basis for Deposition Algorithm

Change in mass flux with distance is proportional to the remaining mass fluxProportionality constant, k, does not vary with heightk varies with vegetative density and wind speedNo further reduction in mass flux downwind of barrier

Dust Transport Across Vegetative Barrier

∫∞

=0

dhx)(h,ME(x)

E(x) = E0 exp (-kx)

Reduction of Plume Mass

0102030405060708090

100

0 20 40 60 80 100Distance (m)

Nea

r-So

urce

Per

cent

Red

uction

k = 0.035 m-1k = 0.0175 m-1k = 0.0035 m-1

Summary

Near-source dust depletion results from Electrostatic agglomeration Deposition on groundcover

Higher winds promote groundcover contactCapture efficiency is similar for PM-10 and PM-2.5Up to 50% plume loss occurs within 20-30 m from the source for tall vegetation

SummaryMitigation Term Offers Decided Advantages for Treatment of Plume Loss

Simpler approachRequires only basic knowledge of groundcover adjacent to sourceConservatism can be added to representation

Effective Method for Dealing with Current Over-Prediction of Dust Impacts

Will provide a source mitigation of at least a factor of 2 over a 50-100 m travel distance